pip install doesn't complete

Trying to retain support for Optimum in NVIDIA garak but this install issue is blocking - can you help?

Here's the trace:

```
$ pip install optimum-nvidia
Collecting optimum-nvidia
Using cached optimum_nvidia-0.1.0b9-py3-none-any.whl.metadata (7.1 kB)
Requirement already satisfied: accelerate>=0.26 in /home/lderczynski/anaconda3/envs/garak/lib/python3.13/site-packages (from optimum-nvidia) (1.11.0)
Requirement already satisfied: datasets>=2.14.0 in /home/lderczynski/anaconda3/envs/garak/lib/python3.13/site-packages (from optimum-nvidia) (3.6.0)
Requirement already satisfied: huggingface-hub>=0.24.0 in /home/lderczynski/anaconda3/envs/garak/lib/python3.13/site-packages (from optimum-nvidia) (0.36.0)
Collecting hf-transfer==0.1.6 (from optimum-nvidia)
Using cached hf_transfer-0.1.6.tar.gz (22 kB)
Installing build dependencies ... done
Getting requirements to build wheel ... done
Installing backend dependencies ... done
Preparing metadata (pyproject.toml) ... done
Collecting mpi4py<4.0.0 (from optimum-nvidia)
Using cached mpi4py-3.1.6.tar.gz (2.4 MB)
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing metadata (pyproject.toml) ... done
Requirement already satisfied: mpmath==1.3.0 in /home/lderczynski/anaconda3/envs/garak/lib/python3.13/site-packages (from optimum-nvidia) (1.3.0)
Collecting numpy<2.0.0,>=1.26.0 (from optimum-nvidia)
Using cached numpy-1.26.4.tar.gz (15.8 MB)
Installing build dependencies ... done
Getting requirements to build wheel ... done
Installing backend dependencies ... done
Preparing metadata (pyproject.toml) ... done
Collecting onnx>=1.17.0 (from optimum-nvidia)
Using cached onnx-1.19.1-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (7.0 kB)
Collecting optimum>=1.21.0 (from optimum-nvidia)
Using cached optimum-2.0.0-py3-none-any.whl.metadata (14 kB)
Requirement already satisfied: setuptools in /home/lderczynski/anaconda3/envs/garak/lib/python3.13/site-packages (from optimum-nvidia) (80.9.0)
Collecting tensorrt-llm==0.16.0 (from optimum-nvidia)
Using cached tensorrt_llm-0.16.0.tar.gz (1.2 kB)
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing metadata (pyproject.toml) ... error
error: subprocess-exited-with-error

× Preparing metadata (pyproject.toml) did not run successfully.
│ exit code: 1
╰─> [59 lines of output]
INFO:wheel-stub:Testing wheel tensorrt_llm-0.16.0-cp310-cp310-linux_x86_64.whl against tag cp310-cp310-linux_x86_64
INFO:wheel-stub:Testing wheel tensorrt_llm-0.16.0-cp310-cp310-win_amd64.whl against tag cp310-cp310-win_amd64
INFO:wheel-stub:Testing wheel tensorrt_llm-0.16.0-cp312-cp312-linux_aarch64.whl against tag cp312-cp312-linux_aarch64
INFO:wheel-stub:Testing wheel tensorrt_llm-0.16.0-cp312-cp312-linux_x86_64.whl against tag cp312-cp312-linux_x86_64
File "/tmp/pip-build-env-l6uzntln/overlay/lib/python3.13/site-packages/wheel_stub/wheel.py", line 249, in download_wheel
return download_manual(wheel_directory, distribution, version, config)
File "/tmp/pip-build-env-l6uzntln/overlay/lib/python3.13/site-packages/wheel_stub/wheel.py", line 185, in download_manual
raise RuntimeError(f"Didn't find wheel for {distribution} {version}")
Traceback (most recent call last):
File "/tmp/pip-build-env-l6uzntln/overlay/lib/python3.13/site-packages/wheel_stub/wheel.py", line 249, in download_wheel
return download_manual(wheel_directory, distribution, version, config)
File "/tmp/pip-build-env-l6uzntln/overlay/lib/python3.13/site-packages/wheel_stub/wheel.py", line 185, in download_manual
raise RuntimeError(f"Didn't find wheel for {distribution} {version}")
RuntimeError: Didn't find wheel for tensorrt-llm 0.16.0

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/home/lderczynski/anaconda3/envs/garak/lib/python3.13/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 389, in
main()
~~~~^^
File "/home/lderczynski/anaconda3/envs/garak/lib/python3.13/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 373, in main
json_out["return_val"] = hook(**hook_input["kwargs"])
~~~~^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/lderczynski/anaconda3/envs/garak/lib/python3.13/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 178, in prepare_metadata_for_build_wheel
whl_basename = backend.build_wheel(metadata_directory, config_settings)
File "/tmp/pip-build-env-l6uzntln/overlay/lib/python3.13/site-packages/wheel_stub/buildapi.py", line 29, in build_wheel
return download_wheel(pathlib.Path(wheel_directory), config_settings)
File "/tmp/pip-build-env-l6uzntln/overlay/lib/python3.13/site-packages/wheel_stub/wheel.py", line 251, in download_wheel
report_install_failure(distribution, version, config, exception_context)
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/pip-build-env-l6uzntln/overlay/lib/python3.13/site-packages/wheel_stub/error.py", line 67, in report_install_failure
raise InstallFailedError(
...<9 lines>...
)
wheel_stub.error.InstallFailedError:
*******************************************************************************

The installation of tensorrt-llm for version 0.16.0 failed.

This is a special placeholder package which downloads a real wheel package
from https://pypi.nvidia.com/. If https://pypi.nvidia.com/ is not reachable, we
cannot download the real wheel file to install.

You might try installing this package via
```
$ pip install --extra-index-url https://pypi.nvidia.com/ tensorrt-llm
```

Here is some debug information about your platform to include in any bug
report:

Python Version: CPython 3.13.9
Operating System: Linux 6.8.0-87-generic
CPU Architecture: x86_64
nvidia-smi command not found. Ensure NVIDIA drivers are installed.

*******************************************************************************

[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

note: This is an issue with the package mentioned above, not pip.
hint: See above for details.

```

Start with the `pip install optimum-nvidia` command and the `tensorrt-llm==0.16.0` metadata failure shown in the trace. Check which Python and platform wheels are available, including the NVIDIA package index, and compare them with the reported CPython 3.13 x86_64 environment. Done means the supported installation path completes or the incompatibility is clearly documented.

Written by the indexing model from the issue text.